Systematic Literature Review of Sentiment Analysis Techniques
Journal of Contemporary Issues in Business and Government,
2021, Volume 27, Issue 1, Pages 506-517
AbstractSentiment analysis has become an important research area that aims to understand people's opinions by analyzing a large size of information. There are two types of sentiment analysis methods: those based on the lexicon and those based on Machine Learning algorithms. Although there are many proposals related to sentiment analysis, there is still a great margin for improving results. The objective of this work is to identify the current state of the latest research related to the analysis of feelings, making use of a framework for the systematic review of the literature, in order to answer the following research questions: RQ1 What are the types of methods used for sentiment analysis? RQ2 What kind of data sources are used to perform sentiment analysis? A crossover analysis of the results was performed. One of the results showed that the most used classifier was Naïve Bayes. Besides, most of the works reviewed used texts extracted from microblogs, web pages, E-Commerce, and other data sources, to perform sentiment analysis.
- Article View: 38
- PDF Download: 64